English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Program translation is an important tool to migrate legacy code in onelanguage into an ecosystem built in a different language. In this work, we arethe first to employ deep neural networks toward tackling this problem. Weobserve that program translation is a modular procedure, in which a sub-tree ofthe source tree is translated into the corresponding target sub-tree at eachstep. To capture this intuition, we design a tree-to-tree neural network totranslate a source tree into a target one. Meanwhile, we develop an attentionmechanism for the tree-to-tree model, so that when the decoder expands onenon-terminal in the target tree, the attention mechanism locates thecorresponding sub-tree in the source tree to guide the expansion of thedecoder. We evaluate the program translation capability of our tree-to-treemodel against several state-of-the-art approaches. Compared against otherneural translation models, we observe that our approach is consistently betterthan the baselines with a margin of up to 15 points. Further, our approach canimprove the previous state-of-the-art program translation approaches by amargin of 20 points on the translation of real-world projects.

Tree-to-tree Neural Networks for Program Translation